Heavy-duty diesel engines are the power horses of the construction industry. Upcoming legislation mandates that nitrogen oxides (NOx) and particulate matter (PM) emissions from these engines be drastically reduced. Through optimal control of the air path management system we have substantially reduced NOx and opacity, a proxy for PM, with respect to conventional control methods.
Our approach to tackling this problem was to use model predictive control of the air path management system of the engine. By including the emissions in the feedback loop and predicting their response subject to changes in speed and torque demand, it was possible to optimally control the variable geometry turbocharger (VGT) and exhaust gas recirculation (EGR) valve. The actuator positions are the result of an optimal control problem where the variables of the system, i.e. intake manifold pressure, compressor air flow, and emissions, are forced to follow a desired trajectory that minimises emissions.
Non-road transient cycles were run in engine test cells showing that this type of optimal control is suitable for engine management and that emissions can be drastically reduced.
This research programme was done in collaboration with Caterpillar Inc and Loughborough University, as part of a larger programme co-funded by the Technology Strategy Board.